Mental illness is a serious health problem associated with poor physical health, functional limitations, and social disadvantage (1–3). Understanding the characteristics of persons with serious mental illness who live in the community can help inform the community-based programs that serve them.

The goal of this study was to describe the epidemiology of serious mental illness in the adult household population by using the National Health Interview Survey (NHIS). Serious mental illness was defined as schizophrenia, bipolar disorder, mania, or psychosis that was diagnosed by a health professional and reported by oneself or by proxy. The NHIS provides the opportunity to duplicate results from smaller studies in a nationally representative sample. Also, because it is a general health survey, the NHIS can provide information on a wider range of health characteristics than more specialized surveys.

Data for this study came from the 2007 NHIS. The NHIS is a cross-sectional survey with a multistage area probability design (www.cdc.gov/nchs/nhis/about_nhis.htm#sample_design). Information is gathered through face-to-face household interviews. Data from the NHIS are weighted to provide estimates for the U.S civilian, noninstitutionalized population.

The final response rate in 2007 for the sample adult interview was 67.8%. Missing data did not exceed 3% for any of the covariates that were used in the analyses except household income. Household income was imputed for about 33% of NHIS respondents and was used to calculate the percentage of persons below the poverty level (cdc.gov/nchs/data/nhis/tecdoc.pdf).

Ten percent of persons with serious mental illness and 1.3% of other adults had proxy respondents. Questions on lifetime serious mental illness were included in the 2007 NHIS sample adult questionnaire. Serious mental illness was reported by answering yes when asked whether one had ever been told by a doctor or other health professional that he or she had schizophrenia, bipolar disorder, or mania or psychosis (three questions).

Homelessness or incarceration was assessed with the question, “Have you ever spent more than 24 hours living on the streets, in a shelter, or in a jail or prison?”

Serious psychological distress was measured by the K6, which asks about the frequency of six symptoms of psychological distress in the past 30 days (4). Items on the K6 are scored on a 5-point Likert scale, with possible scores ranging from 0 to 24. A score of 13 or more indicates serious psychological distress.

The number of chronic conditions was determined by counting heart problems, lung problems, hypertension, diabetes, cancer, and stroke. Lung problems included current asthma, chronic bronchitis, and emphysema. Heart problems included history of angina pectoris, coronary heart disease, heart attack, or other heart condition or disease.

Questions about physical and social limitations asked the respondent to use a 5-point Likert scale to rate the difficulty of doing specific activities. Responses were dichotomized into difficulty (somewhat difficult, very difficult, or can’t do at all) and no difficulty (not at all difficult or a little difficult). Persons who reported not doing the activity were excluded.

Receipt of government or other disability benefits during the past year was assessed with questions from the Family Core Questionnaire. Respondents were asked whether they had received Temporary Assistance to Needy Families, Supplemental Nutrition Assistance Program benefits (food stamps), Social Security Disability Insurance (SSDI), any other disability pension, or Supplemental Security Income (SSI) and whether anyone in the family received governmental assistance with rent. Data on family income, health insurance coverage at the time of interview, past-year hospitalization, living arrangements, and inability to work were also gathered by using the family questionnaire.

The characteristics of adults with and without serious mental illness are presented. Chi square tests were used to detect significant differences between the groups. Because of the multiple comparisons, p≤.01 was used to identify significant results. Percentages are reported as weighted estimates. All analyses utilized SUDAAN, which accounts for the complex design of the survey and calculates appropriate standard errors.

Of the 23,393 adults in the 2007 NHIS sample, 23,374 responded to at least one question concerning serious mental illness and were included in the analytic sample. A total of 514 people (2.2%±.13%) reported lifetime serious mental illness, including 158 (.7%) respondents with mania or psychosis, 150 (.6%) respondents with schizophrenia, and 387 (1.7%) respondents with bipolar disorder. Sixty-five people reported having schizophrenia and bipolar disorder. Most people (N=120) who reported having mania or psychosis also reported having schizophrenia or bipolar disorder.

Adults with serious mental illness were younger, less educated, more likely to be poor and to live alone, and less likely to work than adults without serious mental illness (Table 1). Thirty-six percent of adults with serious mental illness reported having spent at least 24 hours either homeless or in jail during their lifetime. Less than 5% of other adults reported this experience.

Almost 30% of adults with serious mental illness had serious psychological distress within the past 30 days, compared with 2.1% of other adults. They also had higher rates of hypertension and heart and lung problems. Over half of adults with serious mental illness smoked, compared with one-fifth of other adults. Compared with other adults, adults with serious mental illness were twice as likely to report physical limitations, six times as likely to report difficulty relaxing at home, and five times as likely to report difficulties participating in social activities.

Adults with and without serious mental illness were equally likely to have health insurance, but adults with serious mental illness were more likely to have public insurance. Adults with serious mental illness were more likely than other adults to report being unable to afford needed prescription medication, mental health treatment, and dental care. Compared with other adults, adults with serious mental illness had higher health care utilization, including more office visits, emergency room visits, and inpatient hospitalizations. Only 65% of adults with serious mental illness had seen a mental health professional even once in the year prior to the survey.

Over 25% of renters with serious mental illness and less than 10% of renters without serious mental illness lived in a family that received government assistance with rent. More than one-quarter of adults with serious mental illness received food stamps, compared with 4% of other adults. Adults with serious mental illness were also more likely than other adults to receive SSDI, other disability benefits, and SSI. Fifty-three percent of adults with serious mental illness participated in at least one of the six programs examined.

In this study, 2.2% of adults in the household population reported or reported by proxy having received a diagnosis of a serious mental illness. This estimate is likely to be an underestimate of lifetime serious mental illness prevalence in the noninstitutionalized population because the case definition itself probably excludes some cases of serious mental illness. The sensitivity of the definition may be compromised both by the inability to identify persons who have not been given a diagnosis and by respondents’ reluctance to report having a serious mental illness. On the other hand, persons who have never been told by a physician that they have schizophrenia, bipolar disorder, or psychosis or mania are unlikely to report having such conditions, making overestimation of serious mental illness less likely.

As reported in other studies (5), adults with serious mental illness were younger than other adults. However, older adults may be less likely than younger adults to report mental illness because they may associate more stigma with mental illness. Also, there may be survival bias, given studies showing that persons with serious mental illness have higher mortality rates than the general population (6).

This study found that non-Hispanic whites had higher rates of serious mental illness. However, this finding may reflect higher rates of diagnosis among non-Hispanic whites rather than higher rates of serious mental illness. Another possibility is that stigma is more prevalent among other racial and ethnic populations, leading to underreporting by some groups.

Over one-third of the adults with serious mental illness reported a history of homelessness or incarceration. The actual rate of homelessness or incarceration among adults with serious mental illness is likely higher because those who are currently homeless or in a jail or institution are excluded from the NHIS. Studies have documented disproportionate rates of serious mental illness among homeless (7) and incarcerated persons (8).

The physical limitations reported by persons with serious mental illness may be partly due to elevated rates of smoking and chronic disease. Difficulties in relaxing at home or participating in social activities are directly related to symptoms of serious mental illness. The association between mental illness and impairment in many domains, especially the social domain, has been documented in many studies (1,9,10).

Only 65% of adults with serious mental illness had seen a mental health professional even once in the past year. This percentage overestimates the percentage of adults who actually received continuing treatment from a mental health professional. Other studies have estimated that from 39% to 49% of persons with serious mental illness receive specialty mental health treatment (5,11,12).

Despite the fact that 80% of adults with serious mental illness reported having health insurance, one-third of adults with serious mental illness reported not getting a prescription drug and one-quarter reported not getting mental health care because of cost.

To my knowledge, this study is the first to present program participation rates of adults in the household population who have serious mental illness. Over 50% of adults with serious mental illness participated in a government program or were recipients of other disability benefits. The most commonly used programs, with approximately 25% participation, were food stamps and subsidized housing. One cannot determine with the available data whether more adults with serious mental illness qualify for assistance programs than are actually enrolled as beneficiaries.

Because it is a household survey, the NHIS does not interview persons in institutions, but residents of group homes and halfway houses are included. A key limitation of the study is that some misclassification is inherent in self-report. Other limitations include the lack of clinical reappraisal data, the relatively high proportion of proxy respondents, the cross-sectional nature of the survey, and the combining of history of homelessness and of having been in jail in one question. Also, the measure of treatment was weak; information regarding pharmacologic treatment for this population would be useful.

Among the strengths of this study were that the NHIS is nationally representative and thereby generalizable to the noninstitutionalized U.S. population, has a large sample, and includes a wide variety of health topics.

This report describes the characteristics of a nationally representative group of household-dwelling adults with serious mental illness. Household-dwelling adults with serious mental illness are a vulnerable population. Over one-third live in poverty, only one-half are employed, and over one-third have a history of homelessness or of having been in jail. There are large health disparities between persons with and without serious mental illness. Persons with serious mental illness smoke more, exercise less, have higher rates of physical illness and functional limitations, and are more likely to report unmet needs for prescription drugs and mental health care that are due to cost.

Healthy People 2020, the federal government’s health goals for the nation, includes two objectives for persons with serious mental illness—to increase the percentage of persons with serious mental illness who are employed and who receive mental health treatment (healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?Topicid=28). Further studies should aim toward designing interventions that will further the Healthy People 2020 objectives and improve the status of persons with serious mental illness in other areas as well.

The findings and conclusions in this report are those of the author and do not necessarily represent the views of the National Center for Health Statistics, Centers for Disease Control and Prevention.

The author reports no competing interests.